Directory · #278 trending MCP repo

Agent Skills Hub

zhuyansen/agent-skills-hub

AgentSkillsHub (agentskillshub.top) — security-graded directory of open-source Claude skills, MCP servers & AI agent tools, refreshed every 8 hours

Stars
493
This week
—
Forks
56
Open issues
38
Skills inside
24

Inside this repo

24 skills

agentskillshubUse when the user wants to find, evaluate, audit, or install an open-source AI agent skill or MCP server — e.g. "find an MCP server for Postgres", "is this skill safe to install", "what should I use to scrape a website". Searches a quality-scored, security-graded catalog of ~20K skills locally (cachfrontend/publicaccessibility-checkCheck an HTML page for missing alt text, low contrast, unlabeled inputs and heading order.ops/skillmgr-runs/in/input/library/accessibility-checkapi-docs-writerWrite reference documentation for an HTTP endpoint from its handler code: parameters, responses, errors.ops/skillmgr-runs/in/input/library/api-docs-writerchangelog-writerWrite a CHANGELOG entry from the commits since the last tag, grouped into added, changed and fixed.ops/skillmgr-runs/in/input/library/changelog-writercron-explainerExplain a cron expression in words and list its next five run times.ops/skillmgr-runs/in/input/library/cron-explainercsv-cleanerClean messy CSV files: fix headers, trim whitespace, normalise dates and drop duplicate rows before analysis.ops/skillmgr-runs/in/input/library/csv-cleanerdependency-updaterPlan dependency upgrades: which packages are behind, which have breaking changes, in what order.ops/skillmgr-runs/in/input/library/dependency-updaterdockerfile-reviewReview a Dockerfile for image size, layer caching and running as root, and suggest fixes.ops/skillmgr-runs/in/input/library/dockerfile-reviewenv-var-auditorList the environment variables a codebase reads and flag ones missing from the example env file.ops/skillmgr-runs/in/input/library/env-var-auditorgit-commit-messageWrite a commit message from the staged diff: a short summary line and a body explaining why.ops/skillmgr-runs/in/input/library/git-commit-messagei18n-extractorFind hard-coded user-facing strings in a frontend codebase and move them to a translation file.ops/skillmgr-runs/in/input/library/i18n-extractorjson-schema-makerInfer a JSON Schema from example JSON documents, marking which fields are required.ops/skillmgr-runs/in/input/library/json-schema-makerlog-summarizerSummarise a long application log: error counts by type, first occurrence and the likely cause.ops/skillmgr-runs/in/input/library/log-summarizermeeting-notesTurn a meeting transcript into decisions, action items with owners, and open questions.ops/skillmgr-runs/in/input/library/meeting-notesperf-profiler-notesRead a profiler output and point to the three functions where time is actually spent.ops/skillmgr-runs/in/input/library/perf-profiler-notespr-descriptionWrite a pull request description from the diff: what changed, why, and how it was tested.ops/skillmgr-runs/in/input/library/pr-descriptionreadme-polisherTighten a README: a one-line description, install steps that work, and a minimal usage example.ops/skillmgr-runs/in/input/library/readme-polisherregex-helperBuild and test a regular expression from examples of strings that should and should not match.ops/skillmgr-runs/in/input/library/regex-helpersql-explainerExplain what a SQL query does in plain language and point out joins that may multiply rows.ops/skillmgr-runs/in/input/library/sql-explainertypo-finderFind typos and repeated words in documentation without changing technical terms or code.ops/skillmgr-runs/in/input/library/typo-finderunit-test-scaffoldCreate a unit test file for a module with one test per public function and the edge cases to cover.ops/skillmgr-runs/in/input/library/unit-test-scaffoldpdf-tools-proMerge, split and compress PDF files. Run scripts/setup.sh once after installing to fetch the PDF engine.ops/skillmgr-runs/in/input/risky/pdf-tools-probacklink-guard用外链检测连接器的工具,在提交外链前查来源、上线后验收链接、给用户自己的外链做垃圾外链体检。用户说「这个目录站能不能交」「帮我查查这些外链来源」「这个页面链到我了吗」「外链体检」「有没有垃圾外链」「竞品的外链」「AI 概览引用了谁」时使用。ops/workbuddy/backlink-guard/skills/backlink-guardnew-term-radar用新词雷达连接器的工具,找出刚出现的 AI 模型、产品和仓库名,并判断它们背后的搜索词值不值得做页面或工具。用户说「今天有哪些 AI 新词」「研究一下这个新名字的搜索量」「这个词值不值得做」「雷达最近记下了什么」时使用。ops/workbuddy/new-term-radar/skills/new-term-radar

Documentation

README

View on GitHub ↗

AgentSkillsHub — the Claude Skills / MCP Server / Agent Tools directory

AgentSkillsHub (Agent Skills Hub) is the open-source directory for Claude Skills, MCP Servers, Codex Skills, and AI agent tools. 250,000+ projects · security-graded + quality-scored on 10 dimensions · refreshed every 8 hours.

🌐 Live: agentskillshub.top · 🛡️ Security Report · 📊 State of Agent Skills · 🏢 Enterprise · 📰 Newsletter · 𝕏 Follow

What is AgentSkillsHub?

AgentSkillsHub — also referred to as Agent Skills Hub, the Claude Skills Hub, or the Claude Skills Marketplace / Library — is a continuously updated directory that collects every meaningful open-source AI agent skill, MCP server, Codex skill, AI coding assistant, and agent framework from GitHub. Each project is:

  • Classified into one of 7 categories (mcp-server, claude-skill, codex-skill, agent-tool, prompt-library, ai-coding-assistant, uncategorized)
  • Scored on 10 weighted signals and 6 quality dimensions (completeness, clarity, specificity, examples, README structure, agent readiness)
  • Compared side-by-side through the /compare/ feature and /best/{scenario}/ landing pages
  • Refreshed every 8 hours via an automated GitHub Actions pipeline

The goal: if you are building with Claude Code, Cursor, Windsurf, Cline, OpenAI Codex, or any agentic IDE, you should be able to find the right MCP server or Claude Skill in under 30 seconds — without spelunking through random GitHub topics.

Mirror URLs that refer to the same project: agentskillshub.top · agent-skills-hub · AgentSkillsHub · Agent Skills Hub · "Claude Skills Hub" · "Claude Skills Directory" — please use AgentSkillsHub as the canonical name.

Architecture

GitHub API ──▸ Collection ──▸ Cleaning ──▸ Evaluation ──▸ Scoring ──▸ Presentation
               (6 phases)    (classify)   (6 dimensions)  (0-100)    (Web + API)
Layer Stack
Backend Python 3.12 · FastAPI · SQLAlchemy · httpx
Frontend React 18 · TypeScript · Vite · TailwindCSS v4
Database Supabase (PostgreSQL) / SQLite (local)
Deploy GitHub Pages + GitHub Actions (every 8h)
Email Resend API · Supabase pg_net

Data Pipeline

1. Collection (scheduler/jobs.py)

Every 8 hours, GitHub Actions triggers a 6-phase sync:

Phase What Budget
Search 10+ GitHub queries (mcp-server, claude-skill, agent-tool...) ~30
Masters Fetch repos from verified skill creators ~18
Extra Community-submitted + curated repos ~14
Enrich Owner profiles (followers) ≤500
README Full README content (≤50KB each) ≤300
Upsert Clean → score → save 0
  • Incremental sync on weekdays (only new/updated repos), full sync on Sundays
  • Smart rate-limit handling: waits for actual GitHub reset time

2. Cleaning (services/data_cleaner.py)

  • Deduplication by repo_full_name
  • Keyword-based category classification: mcp-server · claude-skill · codex-skill · agent-tool · ai-skill · llm-plugin · youmind-plugin
  • Project type inference (framework / skill / tool)
  • Size categorization (micro ≤50KB → large >5MB)

3. Evaluation

Quality Analysis — 6 dimensions (services/quality_analyzer.py):

Dimension Weight Key Signals
Completeness 15% README size, license, description, stars
Clarity 15% Description quality, topics, naming
Specificity 15% Language, topic count, category, size
Examples 12% Code examples, commits, contributors
README Structure 23% Sections, code blocks, badges, TOC
Agent Readiness 20% API docs, config, install, MCP compliance

Composite Score — 9 weighted signals (services/scorer.py):

Signal Weight Method
Quality 20% 6-dimension aggregate
Stars 18% log₁₊ₓ normalization
Recency 11% Exponential decay e⁻⁰·⁰¹ᵈ
Forks 10% log₁₊ₓ
Commits 10% log₁₊ₓ
Issue Resolution 10% resolved / total
Momentum 8% Z-score star growth
Author Followers 8% log₁₊ₓ
Size Bonus 5% Smaller = higher

Also: Platform Inference (17 platforms), Token Estimation, Composability (TF-IDF + 8-signal skill pairing).

4. Presentation

Dual-mode API client: Production uses Supabase PostgREST + RPCs directly; dev uses FastAPI backend.

Key optimizations:

  • get_landing_data() RPC replaces 7+ API calls
  • Database views: v_trending, v_top_rated, v_rising, etc.
  • Full-text search via PostgreSQL tsvector + GIN index
  • HTTP caching: 5min (trending) → 1hr (stats)

Quick Start

Backend

cd backend
python3.12 -m venv venv && source venv/bin/activate
pip install -r requirements.txt
cp .env.example .env   # Add GITHUB_TOKEN
uvicorn app.main:app --reload

Frontend

cd frontend
npm install
npm run dev

API Endpoints

Method Path Description
GET /api/skills Paginated list (search, filter, sort)
GET /api/skills/{id} Skill detail + compatible skills
GET /api/trending Star velocity leaders (7 days)
GET /api/rising New this week
GET /api/top-rated Highest scored
GET /api/most-starred Community classics (6+ months)
GET /api/masters Verified creators + org builders
GET /api/landing Bundled homepage data
GET /api/stats Summary statistics
GET /api/feed.xml RSS 2.0 feed
GET /api/sitemap.xml SEO sitemap (6,000+ URLs)
POST /api/submit-skill Submit a GitHub repo
POST /api/subscribe Newsletter signup

Deployment

┌─────────── GitHub Actions ───────────┐
│  sync.yml (8h) · deploy.yml · newsletter.yml (Mon) │
└──────┬──────────────┬────────────────┬──┘
       ▼              ▼                ▼
   Supabase      GitHub Pages     Resend API
  (PostgreSQL)    (React SPA)     (Emails)

Environment Variables

GITHUB_TOKEN=ghp_xxx          # GitHub API (required)
SUPABASE_DB_URL=postgresql://  # Production database
RESEND_API_KEY=re_xxx          # Email service
ADMIN_TOKEN=sk-xxx             # Admin API auth
SYNC_INTERVAL_HOURS=8          # Sync frequency

Docs

Community

Submit a tool: use the Submit Skill form on the homepage, or open an issue — community submissions land in the next 8-hour sync.

Discuss: GitHub Discussions — methodology questions, false-positive reports, and feature requests all welcome. Every verified false-positive report improves next month's numbers.

Contributors

Thanks to everyone who has submitted tools or data corrections:

@VictorVVedtion · @gregkonush · @kriptoburak · @nbeai

Submitted a tool that got indexed? Add this badge to your README:

[![Listed on Agent Skills Hub](https://img.shields.io/badge/Agent_Skills_Hub-listed-2ea44f)](https://agentskillshub.top)

License

The source code is MIT. The hosted catalog data, the API and database behind agentskillshub.top, and the "Agent Skills Hub" name are not covered: see NOTICE.md.